Support is requested for 3 years to conduct secondary analyses of transitions in drug addiction, using data from the NIDA-funded Harvard Drug Study (HDS), in which 8169 members of the Vietnam Era Twin Registry (VETR) (a large general population sample of male same-sex twins who served in the military during the Vietnam era) were interviewed by telephone with a structured psychiatric diagnostic interview. Covered in the interview were DSM-III-R lifetime criteria for dependence and abuse on 5 illicit and 2 licit drug classes, as well as other psychiatric disorders. The availability of ages of onset and offset for each drug-class specific symptom of addiction in these data make them uniquely suitable for a method of studying transitions to drug addiction that we have already used successfully with alcohol symptom data. The twin structure of the data permits study of the effect of genetic factors on occurrence and transitions to addiction on specific drugs. Specifically we will: 1) explore the nosology of drug-class specific addictions and investigate the role of genetic factors in the classifications, using latent structure techniques like latent class analysis; 2) investigate transitions in addiction in both epidemiologic and genetic frameworks, using a) a more conventional approach that relies on drug use patterns (e.g. initiation, regular and problem use), as well as b) a novel approach based on person-years of addiction symptoms, developing a method for incorporating symptom offsets; 3) study factors (including use of other substances, psychiatric comorbidity, social characteristics, family history) that may promote, accelerate, or inhibit transitions; and 4) conduct, wherever possible, cross-study analyses using other sources of data, especially data from the Vietnam Era Study (VES), National Comorbidity Survey (NCS), National Longitudinal Alcohol Epidemiology Survey (NLAES), and the Collaborative Study on the Genetics of Alcoholism (COGA), in order to test the replicability of the findings. We expect our proposed analyses will contribute to an understanding of transitions in addiction to specific types of drugs. Ultimately, our research may suggest certain points in transitions in specific drug addictions that might be amenable to preventive interventions.